The morbidity of cardiovascular disease increasingly rises, which makes great impact upon people’s health and life. Electrocardiogram (ECG) beat classification is of great significance to clinical diagnosis of cardiovascular diseases. Traditional ECG signal classification algorithm relies heavily on the accuracy of feature extraction or increases the complexity of the calculation process by means of the correlation characteristic coefficient transformation, which results in that the ECG beat classification effect is still not satisfactory. Aimed at this problem, a novel method based on convolution neural network (CNN) is presented in this paper. First, ECG signal is preprocessed to suppress the noise and to locate the R peaks, and five kinds of ECG beat waveform data are obtained. Then taking ECG beat sampling points as input, four layers of one-dimensional CNN are constructed for feature extraction and classification. Finally, experimental verification is carried out on the data from MIT-BIH database, and the accuracy of recognition and classification of the presented method reaches 99.10%. Comparison with the methods based on artificial features, this method shows better performance, which avoids serious dependence on the accuracy of feature extraction, skips the steps of feature extraction and selection, and reduces the complexity of computational process.

BACKGROUND: Ventricular repolarization instabilities have been documented to be closely linked to arrhythmia development. The electrocardiogram (ECG) ST interval can be used to measure ventricular repolarization. Analyzing the duration variation of the ST intervals can provide new information about the arrhythmogenic vulnerability. OBJECTIVE: In this work, we propose a new method based on mean instantaneous frequency (IF) of the ST intervals to quantitatively evaluate the risk of sudden cardiac deaths (SCDs). METHODS: Two spectral bands, i.e. the low-frequency band (LF, 0–0.15 Hz) and the high-frequency band (HF, 0.15–0.5 Hz), are considered in this paper. Based on IF estimates, the ECG recordings from three MIT-BIH databases that represent different risk levels of SCD occurrence are used, and their mean IFs in the LF and HF bands are calculated. RESULTS: The statistical results show that healthy subjects have a higher mean IF in the HF band and a lower mean IF in the LF band. The experimental results are the opposite for patients with malignant ventricular arrhythmia. CONCLUSION: The proposed mean IF can represent an indirect measure of intrinsic ventricular repolarization instability and can mark cardiac instability associated with SCDs.

In recent years, the number of cardiac disease patients has been increasing. Modern medical research has shown that the complexity of electrocardiogram (ECG) signals is related to cardiovascular diseases. This paper investigates the difference in complexity of ECG data from the people with different cardiovascular diseases, such as atrial fibrillation (AF), ventricular arrhythmia (VA) and congestive heart failure (CHF). The empirical mode decomposition (EMD) and multiscale entropy method are used to analyze the ECG data, and a mathematical model established by a support vector machine is used to identify different diseases. The accuracy recognition rate of the AF recognition is 96.25%, and that of the CHF and VA reach 90.26% and 92.20%, respectively. The experimental results show that the recognition method proposed in this paper is successful.